Semiconductor nanocrystals are typically tiny crystals of II-VI, III-V, IV-VI materials that have a diameter between 1 nanometer (nm) and 20 nm. In the strong confinement limit, the physical diameter of the nanocrystal is smaller than the bulk excitation Bohr radius causing quantum confinement effects to predominate. In this regime, the nanocrystal is a 0-dimensional system that has both quantized density and energy of electronic states where the actual energy and energy differences between electronic states are a function of both the nanocrystal composition and physical size. Larger nanocrystals have more closely spaced energy states and smaller nanocrystals have the reverse. Because interaction of light and matter is determined by the density and energy of electronic states, many of the optical and electric properties of nanocrystals can be tuned or altered simply by changing the nanocrystal geometry (i.e. physical size).
Single nanocrystals or monodisperse populations of nanocrystals exhibit unique optical properties that are size tunable. Both the onset of absorption and the photoluminescent wavelength are a function of nanocrystal size and composition. The nanocrystals will absorb all wavelengths shorter than the absorption onset, however, photoluminescence will always occur at the absorption onset. The bandwidth of the photoluminescent spectra is due to both homogeneous and inhomogeneous broadening mechanisms. Homogeneous mechanisms include temperature dependent Doppler broadening and broadening due to the Heisenburg uncertainty principle, while inhomogeneous broadening is due to the size distribution of the nanocrystals. The narrower the size distribution of the nanocrystals, the narrower the full-width half max (FWHM) of the resultant photoluminescent spectra. In 1991, Brus wrote a paper reviewing the theoretical and experimental research conducted on colloidally grown semiconductor nanocrystals, such as cadmium selenide (CdSe) in particular. Brus L., Quantum Crystallites and Nonlinear Optics, Applied Physics A, 53 (1991)). That research, precipitated in the early 1980's by the likes of Efros, Ekimov, and Brus himself, greatly accelerated by the end of the 1980's as demonstrated by the increase in the number of papers concerning colloidally grown semiconductor nanocrystals.
Quantum yield (i.e. the percent of absorbed photons that are reemitted as photons) is influenced largely by the surface quality of the nanocrystal. Photoexcited charge carriers will emit light upon direct recombination but will give up the excitation energy as heat if photon or defect mediated recombination paths are prevalent. Because the nanocrystal may have a large surface area to volume ratio, dislocations present on the surface or adsorbed surface molecules having a significant potential difference from the nanocrystal itself will tend to trap excited state carriers and prevent radiactive recombination and thus reduce quantum yield. It has been shown that quantum yield can be increased by removing surface defects and separating adsorbed surface molecules from the nanocrystal by adding a shell of a semiconductor with a wider bulk bandgap than that of the core semiconductor.
Inorganic colloids have been studied for over a century ever since Michael Faraday's production of gold sols in 1857. Rossetti and Brus began work on semiconductor colloids in 1982 by preparing and studying the luminescent properties of colloids consisting of II-VI semiconductors, namely cadmium sulfide (CdS). (Rossetti, R.; Brus L., Electron-Hole Recombination Emission as a Probe of Surface Chemistry in Aqueous CdS Colloids, J. Phys. Chem., 86, 172 (1982)). In that paper, they describe the preparation and resultant optical properties of CdS colloids, where the mean diameter of the suspended particles is greater than 20 nm. Because the sizes of the particles were greater than the exaction Bohr radius, quantum confinement effects that result in the blue shifting of the fluorescence peak was not observed. However, fluorescence at the bulk bandedge energies were observed and had a FWHM of 50-60 nm.
CdS colloids exhibiting quantum confinement effects (blue shifted maxima in the absorption spectra) were being prepared since 1984. (Fotjik A., Henglein A., Ber. Bunsenges. Phys. Chem., 88, (1984); Fischer C., Fotjik A., Henglein A., Ber. Bunsenges. Phys. Chem., (1986)). In 1987, Spanhel and Henglein prepared CdS colloids having mean particle diameters between 4 and 6 nm. (Spanhel L., Henglein A., Photochemistry of Colloidal Semiconductors, Surface Modification and Stability of Strong Luminescing CdS Particles, Am. Chem. Soc., 109 (1987)). The colloids demonstrated quantum confinement effects including the observation of size dependent absorption maxima (first exciton peaks) as well as size dependent fluorescent spectra. The colloids were prepared by bubbling a sulphur containing gas (H2S) through an alkaline solution containing dissolved cadmium ions. The size and resultant color (of the fluorescence) of the resultant nanocrystals were dependent upon the pH of the solution. The colloids were further modified or “activated” by the addition of cadmium hydroxide to the solution that coated the suspended nanocrystals. The resultant core-shell nanocrystals demonstrated that the quantum yield of the photoluminescence was increased from under 1% to well over 50% with a FWHM of the photoluminescent spectra under 50 nm for some of the preparations.
Kortan and Brus developed a method for creating CdSe coated zinc sulphide (ZnS) nanocrystals and the opposite, zinc sulphide coated cadmium selenide nanocrystals. (Kortan R., Brus L., Nucleation and Growth of CdSe on ZnS Quantum Crystallite Seeds, and Vice Versa, in Inverse Micelle Media, J. Am. Chem. Soc., 112 (1990)). The preparation grew ZnS on CdSe “seeds” using a organometallic precursor-based reverse micelle technique and kept them in solution via an organic capping layer (thiol phenol). The CdSe core nanocrystals had diameters between 3.5 and 4 nm and demonstrated quantum confinement effects including observable exciton absorption peaks and blue shifted photoluminescence. Using another preparation, CdSe cores were coated by a 0.4 nm layer of ZnS. The photoluminescence spectra of the resultant core-shell nanocrystals indicates a peak fluorescence at 530 nm with an approximate 40-45 nm FWHM.
Murray and Bawendi developed an organometallic preparation capable of making CdSe, CdS, and CdTe nanocrystals. (Murray C., Norris D., Bawendi M., Synthesis and Characterization of Nearly Monodisperse CdE (E=S, Se, Te) Semiconductor Nanocrystallites, J. Am. Chem. Soc., 115, (1993)). This work, based on the earlier works of Brus, Henglein, Peyghambarian, allowed for the growth of nanocrystals having a diameter between 1.2 nm and 11.5 nm and with a narrow size distribution (<5%). The synthesis involved a homogeneous nucleation step followed by a growth step. The nucleation step is initiated by the injection of an organometallic cadmium precursor (dimethyl cadmium) with a selenium precursor (TOPSe-TriOctylPhosphine Selenium) into a heated bath containing coordinating ligands (TOPO-TriOctylPhosphineOxide). The precursors disassociate in the solvent, causing the cadmium and selenium to combine to form a growing nanocrystal. The TOPO coordinates with the nanocrystal to moderate and control the growth. The resultant nanocrystal solution showed an approximate 10% size distribution, however, by titrating the solution with methanol the larger nanocrystals could be selectively precipitated from the solution thereby reducing the overall size distribution. After size selective precipitation, the resultant nanocrystals in solution were monodisperse (capable of reaching a 5% size distribution) but were slightly prolate (i.e. nonspherical having an aspect ratio between 1.1 and 1.3). The photoluminescence spectra show a FWHM of approximately 30-35 nm and a quantum yield of approximately 9.6%.
Katari and Alivisatos slightly modified the Murray preparation to make CdSe nanocrystals. (Katari J., Alivisatos A., X-ray Photoelectron Spectroscopy of CdSe Nanocrystals with Applications to Studies of the Nanocrystal Surface, J. Phys. Chem., 98 (1994)). They found that by substituting the selenium precursor TOPSe with TBPSe (TriButylPhosphineSelenide), nanocrystals were produced that were monodisperse without size selective precipitation, were crystalline, and spherical. The nanocrystals were size tunable from 1.8 nm to 6.7 nm in diameter and had an exciton peak position ranging from 1.9-2.5 eV (corresponding to 635-496 nm wavelength). Like the Murray paper, TOPO was used as the coordinating ligand.
Hines and Guyot-Sionest developed a method for synthesizing a ZnS shell around a CdSe core nanocrystal. (Hines et al., “Synthesis and Characterization of strongly Luminescing ZnS capped CdSe Nanocrystals”; J. Phys. Chem., 100:468-471 (1996)). The CdSe cores, having a monodisperse distribution between 2.7 nm and 3.0 nm (i.e. 5% size distribution with average nanocrystal diameter being 2.85 nm), were produced using the Katari and Alivisatos variation of the Murray synthesis. The photoluminescence spectra of the core shows a FWHM of approximately 30 nm with a peak at approximately 540 nm. The core CdSe nanocrystals were separated, purified, and resuspended in a TOPO solvent. The solution was heated and injected with zinc and sulphur precursors (dimethyl zinc and (TMS)2S) to form a ZnS shell around the CdSe cores. The resultant shells were 0.6±0.3 nm thick, corresponding to 1 -3 monolayers. The photoluminescence of the core-shell nanocrystals had a peak at 545 nm, FWHM of 40 nm, and a quantum yield of 50%.
One exemplary problem associated with known colloidal shell synthesis methods is that they are limited in regard to the growth of high quality III-V semiconductor nanocrystals due, at least in part, to the covalent nature of the material. Past attempts at growing III-V semiconductor nanocrystals have resulted in quantum yields of approximately 10%. Micic et al., J. Phys. Chem. B. 2000, 104, 12149 -12156. It has been shown using a slow growth technique that indium phosphate (InP) core nanocrystals with a shell of zinc sulphide (ZnS) may result in quantum yields of 20% but this process takes days to complete. Haubold et al., ChemPhysChem 2001, 5, 331. Many applications of nanocrystal complexes, such as LEDs, inks, and pigments, require higher luminescent quantum yields.
Additionally, the growth of III-V nanocrystals is difficult because of the need for vacuum growth and strict humidity conditions. Etching has shown to improve the quantum yield of III-V materials however, the etching process introduces problems when bonding ligands to the surface of the nanocrystals. This causes poor colloidal stability and makes the complexes incompatible for a variety of applications. In addition, the etched surface does not provide sufficient electronic passivation. As a result, electronic interaction between a matrix material and the nanocrystal may result in fluorescence quenching.
Additionally, etching relies on photoexcitation, which requires dilute mixtures. This is not conducive to the scale up and production of large quantities of III-V semiconductor nanocrystal complexes. By readily losing their fluorescence, these quantum dot materials are simply unsuitable for most applications.
Thus, it is necessary to develop a stable III-V semiconductor nanocrystal complex that is brightly fluorescing, and soluble in most common solvents.